[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

closing upstream bugs with debian/changelog (was: Accepted gimp 2.0.0-1 (i386 source all))

On Sat, Apr 03, 2004 at 08:58:24AM +1000, Herbert Xu wrote:
> Branden Robinson <branden@debian.org> wrote:
> > 
> > On Thu, Apr 01, 2004 at 08:48:17PM +1000, Herbert Xu wrote:
> > [...]
> >> >> >  gimp (2.0.0-1) unstable; urgency=low
> >> >> >  .
> >> >> >    * New upstream release
> >> >> >      - Closes: #21393, #29375, #64000, #70461, #101266, #108834, #200357
> > [...]
> >> If these bugs were fixed by changes new in this upstream release, then
> >> he'd have been perfectly right in closing them here.
> > 
> > Only if every single one of those was "new upstream version available,
> > please package it".
> > 
> > Otherwise, the reason for closing the bug should be specifically
> > explained.
> Nope, upstream changes go into the upstream changelog.  debian/changelog
> should only contain Debian changes.

If the Debian changelog is going to be used to close bugs in the Debian
BTS, it needs to explain why those bugs are being closed.  The above
quoted snippet of Debian changelog, in which (as it turns out) a lot of
bugs were closed even though the new upstream release had little to
nothing to do with those bugs' resolution, is a good example why the
maintainer/uploader's understanding of why the bug is closed needs to be
stated, not left to interpretation.

Permit me to offer an example:

  * new upstream release
    - XFree86 X server now supports ATI Radeon Mobility (Closes: #196810)
    - XFree86 X server now supports Intel i845G (Closes: #184322)
    - XFree86 X server now supports Intel i865G (Closes: #221686)
    - XFree86 X server now supports SiS 650, 651, and 740 (Closes: #183619)
    - XTerm allows window title reporting, disabled by default, to be enabled
      at the user's option via the allowWindowOps X resource (Closes: #200857)
    - Wacom driver has improved support for Graphire 2 tablets
      (Closes: #165849)
    - Wacom driver now supports absolute motion on Graphire tablets
      (Closes: #188842)
    - extraneous tokens at end of MakeSimpleDoc #ifdef in X11.tmpl removed
      (Closes: #196058)
    - XVideo supported at high resolutions on ATI Rage 128 Mobility LF
      (Closes: #176014)
    - Xinerama header file now usable by C++ code (Closes: #224757)
    - XFree86 X server has improved Xv support for ATI Radeon chipsets
      (Closes: #228408)
    - S3 driver now properly restores video mode on VT switch back to X on
      Trio 32/64 rev 67 (Closes: #164780)
    - ATI/Radeon driver now properly unblanks LCD on Radeon Mobility M7 LW rev
      0 (Closes: #169308)
    - Trident driver now works properly with TGUI 96xx rev 211
      (Closes: #170396)
    - xwd manpage no longer cross-references nonexistent xpr manpage
      (Closes: #104241)

In the above, anyone who cares to read the entry can clearly understand
the justification for the bug's closure, and the maintainer/uploader
clearly not simply closing irrelevant bugs with the Debian changelog out
of laziness.

"Herbert Xu would never do such a thing" is not a valid defense.  Our
changelogs need to be accessible to those who are not aware of your
reputation within the Debian Project.

Good changelog practices speak for themselves, and do not call upon the
reader to check the name at the end of the changelog entry to weigh the
veracity or applicability of the text.

Documentation is superior to folklore.

G. Branden Robinson                |    I reverse the phrase of Voltaire,
Debian GNU/Linux                   |    and say that if God really existed,
branden@debian.org                 |    it would be necessary to abolish
http://people.debian.org/~branden/ |    him.             -- Mikhail Bakunin

Attachment: signature.asc
Description: Digital signature

Reply to: